ei-tui 1.6.1 → 1.6.3

package/README.md

@@ -1,12 +1,12 @@
 # Ei
 
-A local-first AI companion system with persistent personas and coding tool integrations (OpenCode, Claude Code, Cursor).
+A local-first AI companion system with persistent personas and coding tool integrations (OpenCode, Claude Code, Cursor, Codex).
 
 You can access the Web version at [ei.flare576.com](https://ei.flare576.com).
 
 You can run the local version via `bunx ei-tui` — no install needed, always current (see [### TUI](#tui) for details).
 
-If you're here to give your coding tools (OpenCode, Claude Code, Cursor) persistent memory, jump over to [TUI README.md](./tui/README.md) to learn how to get information _into_ Ei, and [CLI README.md](./src/cli/README.md) to wire up automatic context injection so your agents receive relevant memory before every message — no tool calls required.
+If you're here to give your coding tools (OpenCode, Claude Code, Cursor, Codex) persistent memory, jump over to [TUI README.md](./tui/README.md) to learn how to get information _into_ Ei, and [CLI README.md](./src/cli/README.md) to wire up MCP/context injection so your agents can read relevant memory when they need it.
 
 ## What Does "Local First" Mean?
 
@@ -83,7 +83,7 @@ Ei can operate with three types of input, and three types of output.
                        ^
                     Sessions
                        |
-          [OpenCode / Claude Code / Cursor]
+          [OpenCode / Claude Code / Cursor / Codex]
 ```
 
 ```
@@ -91,7 +91,7 @@ Ei can operate with three types of input, and three types of output.
                           |
                        CLI Data
                           v
-                      [OpenCode]
+              [OpenCode / Claude Code / Cursor / Codex]
 ```
 
 Optionally, users can opt into a server-side data sync. This is ideal for users who want to use multiple devices or switch between TUI and Web throughout the day. All data is encrypted _before_ being sent to the server, using a key that only the user can generate (your `username` and `passphrase` never leave your device - I couldn't decrypt your data if I wanted to).
@@ -129,7 +129,7 @@ More information (including commands) can be found in the [TUI Readme](tui/READM
 
 Ei can import sessions from your coding tools and extract what you've been working on — pulling out facts, topics, and context that persist across sessions. Enable any combination; they work independently and feed into the same knowledge base.
 
-All three integrations are enabled via `/settings` in the TUI.
+All four integrations are enabled via `/settings` in the TUI.
 
 #### OpenCode
 
@@ -165,6 +165,21 @@ Reads from Cursor's SQLite databases:
 
 All sessions map to a single "Cursor" persona.
 
+#### Codex
+
+```yaml
+codex:
+  integration: true
+```
+
+Reads from Codex's local state database and rollout JSONL files:
+- `~/.codex/state_*.sqlite`
+- `~/.codex/sessions/`
+
+All sessions map to a single "Codex" persona. Codex may store thread-level agent metadata for custom agents, but rollout messages do not reliably expose per-message sub-agent speaker identity yet. Tool calls, prompt scaffolding, and token-count events are stripped — only visible user/agent messages are imported.
+
+Codex can also read Ei's knowledge back out. Run `ei --install` to register the Ei MCP server and install a Codex `UserPromptSubmit` hook that injects relevant memory before each message, using your current prompt plus recent Codex transcript context when available.
+
 ---
 
 Sessions are processed oldest-first, one per queue cycle, so Ei won't overwhelm your LLM provider on first run. See [TUI Readme](tui/README.md)
@@ -233,7 +248,7 @@ Personas can use tools. Not just read-from-memory tools — *actual* tools. Web
 |------|-------------|
 | `find_memory` | Semantic search of your personal memory — facts, traits, topics, people, quotes. Personas call this automatically when the conversation touches something they might know about you. Supports the `persona` filter to scope results to what a specific persona has learned. |
 | `fetch_memory` | Full-record lookup for a specific human entity (Fact, Topic, Person, or Quote) by ID. Use after `find_memory` to retrieve complete details. |
-| `fetch_message` | Retrieve a specific message by ID with optional surrounding context. Searches persona conversations and room messages. |
+| `fetch_message` | Retrieve a specific message by fully-qualified ID with optional surrounding context. Searches Ei conversations and supported external coding-session stores. |
 | `file_read` | Read a file from your local filesystem *(TUI only)* |
 | `list_directory` | Explore folder structure *(TUI only)* |
 | `directory_tree` | Recursive directory tree *(TUI only)* |

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "ei-tui",
-  "version": "1.6.1",
+  "version": "1.6.3",
   "author": "Flare576",
   "repository": {
     "type": "git",

package/src/cli/README.md

@@ -15,8 +15,8 @@ ei --recent                            # Most recently mentioned items (no query
 ei --persona "Beta" --recent           # Most recently mentioned items Beta has learned
 ei --id <id>                   # Look up entity by ID — or fetch a message by FQ ID
 echo <id> | ei --id            # Look up entity by ID from stdin
-ei --install                   # Wire Ei into Claude Code, Cursor, and OpenCode (MCP + hooks + persona plugin)
-ei mcp                         # Start the Ei MCP stdio server (for Cursor/Claude Desktop)
+ei --install                   # Wire Ei into Claude Code, Cursor, Codex, and OpenCode (MCP + hooks + persona plugin)
+ei mcp                         # Start the Ei MCP stdio server (for Claude Code/Cursor/Codex)
 ```
 
 Type aliases: `fact`, `person`, `topic`, `quote`, `persona` all work (singular or plural).
@@ -35,6 +35,8 @@ It also resolves fully-qualified message IDs from any supported integration, ret
 ei --id "opencode:jeremys-macbook-pro:ses_38a7...:msg_c75b..."
 ei --id "claudecode:my-machine:session-uuid:message-uuid"
 ei --id "cursor:my-machine:composer-uuid:bubble-uuid"
+ei --id "codex:my-machine:thread-uuid:evt_42"
+ei --id "pi:my-machine:session-uuid:session-uuid/entry-id"
 ```
 
 Quotes surfaced by `ei_search` include a `message_id` field in this format — pipe it to `ei --id` to read the original conversation.
@@ -47,15 +49,17 @@ Quotes surfaced by `ei_search` include a `message_id` field in this format — p
 ei --install
 ```
 
-This registers Ei with Claude Code, Cursor, and OpenCode — MCP server config, context injection hooks, and (for OpenCode) a persona identity plugin so agents know who they are before the first message:
+This registers Ei with Claude Code, Cursor, Codex, and OpenCode — MCP server config, context injection hooks where supported, and (for OpenCode) a persona identity plugin so agents know who they are before the first message:
 
 | Tool | MCP | Context Hook | Persona Plugin |
 |------|-----|-------------|----------------|
 | **Claude Code** | `~/.claude.json` | `~/.claude/settings.json` (`UserPromptSubmit`) + `~/.claude/hooks/ei-inject.ts` | — |
 | **Cursor** | `~/.cursor/mcp.json` | `~/.cursor/hooks.json` (`beforeSubmitPrompt`) + `~/.cursor/hooks/ei-inject.sh` | — |
+| **Codex** | `~/.codex/config.toml` via `codex mcp add ei` | `~/.codex/hooks.json` (`UserPromptSubmit`) + `~/.codex/hooks/ei-inject.ts` | Local Codex agent plugin if installed separately |
 | **OpenCode** | manual (see below) | Via Oh My OpenCode compatibility layer (reads `~/.claude/settings.json`) | `~/.config/opencode/plugins/ei-persona.ts` |
+| **Pi / OMP** | — (tools registered as native Pi extension) | `~/.pi/agent/extensions/ei-integration.ts` (Pi) or `~/.omp/agent/extensions/ei-integration.ts` (OMP) | — |
 
-**Context hook**: fires before every message, searches Ei for topics relevant to what you just asked, injects them silently. No tool call required.
+**Context hook**: fires before every message, searches Ei for relevant memory, and injects it silently. No tool call required.
 
 **Persona plugin** (OpenCode + [Oh My OpenCode](https://github.com/code-yeongyu/oh-my-opencode) only): injects the agent's Ei relationship record directly into the system prompt at session start — traits, working style, shared context. The agent knows who it is *to you* before it reads a word of your message.
 
@@ -80,7 +84,7 @@ Restart your agent tool after changes to activate.
 
 ### MCP Server
 
-Claude Code and Cursor call `ei mcp` to start the MCP stdio server. You can run it directly to test:
+Claude Code, Cursor, and Codex call `ei mcp` to start the MCP stdio server. You can run it directly to test:
 
 ```sh
 ei mcp
@@ -97,7 +101,7 @@ The `ei_search`, `ei_lookup`, and `ei_fetch_message` MCP tools are still availab
 
 ## MCP Tools Reference
 
-The MCP server exposes these tools to Claude Code, Cursor, and OpenCode:
+The MCP server exposes these tools to Claude Code, Cursor, Codex, and OpenCode:
 
 | Tool | Description |
 |------|-------------|
@@ -112,7 +116,7 @@ The MCP server exposes these tools to Claude Code, Cursor, and OpenCode:
 | `query` | string (optional) | Search text. Omit to browse by recency. |
 | `type` | enum (optional) | `facts` \| `people` \| `topics` \| `quotes` \| `personas` — omit for balanced results across all types |
 | `persona` | string (optional) | Persona display_name to scope results to what that persona has learned |
-| `source` | string (optional) | Prefix match against source identifiers (e.g. `opencode`, `cursor:my-machine`) |
+| `source` | string (optional) | Prefix match against source identifiers (e.g. `opencode`, `cursor:my-machine`, `codex:my-machine`) |
 | `limit` | number (optional) | Max results, default 10 |
 | `recent` | boolean (optional) | Sort by most recently mentioned instead of relevance |
 

package/src/cli/mcp.ts

@@ -37,7 +37,7 @@ export function createMcpServer(): McpServer {
           .string()
           .optional()
           .describe(
-            "Filter to entities from a specific source. Prefix match against namespaced source identifiers (e.g. 'cursor', 'opencode', 'opencode:my-machine', 'opencode:my-machine:ses_abc123')."
+            "Filter to entities from a specific source. Prefix match against namespaced source identifiers (e.g. 'cursor', 'codex', 'opencode', 'opencode:my-machine', 'codex:my-machine:thread-id')."
           ),
         limit: z
           .number()
@@ -106,7 +106,7 @@ export function createMcpServer(): McpServer {
           .string()
           .optional()
           .describe(
-            "Filter to entities from a specific source. Prefix match against namespaced source identifiers (e.g. 'cursor', 'opencode', 'opencode:my-machine', 'opencode:my-machine:ses_abc123'). If the entity does not match, returns not found."
+            "Filter to entities from a specific source. Prefix match against namespaced source identifiers (e.g. 'cursor', 'codex', 'opencode', 'opencode:my-machine', 'codex:my-machine:thread-id'). If the entity does not match, returns not found."
           ),
       },
     },
@@ -134,7 +134,7 @@ export function createMcpServer(): McpServer {
     "ei_fetch_message",
     {
       description:
-        "Retrieve a specific message by its fully-qualified ID, with optional surrounding conversation context. Use when ei_search returns a quote with a message_id and you want to read the original exchange. The 'before' and 'after' parameters expand the context window in either direction (default 0). Accepts IDs from any integrated source: 'ei:uuid' searches Ei state, 'opencode:machine:session:id' queries OpenCode SQLite, 'claudecode:...' scans Claude Code JSONL files, 'cursor:...' reads the Cursor DB.",
+        "Retrieve a specific message by its fully-qualified ID, with optional surrounding conversation context. Use when ei_search returns a quote with a message_id and you want to read the original exchange. The 'before' and 'after' parameters expand the context window in either direction (default 0). Accepts IDs from any integrated source: 'ei:uuid' searches Ei state, 'opencode:machine:session:id' queries OpenCode SQLite, 'claudecode:...' scans Claude Code JSONL files, 'cursor:...' reads the Cursor DB, and 'codex:...' reads Codex rollout history.",
       inputSchema: {
         id: z.string().describe("The ID of the message to retrieve"),
         before: z

package/src/cli/retrieval.ts

@@ -499,6 +499,50 @@ export async function resolveExternalMessage(
       }
     }
 
+    case "codex": {
+      if (parsed.machine !== getMachineId()) {
+        return { error: `Message is from machine '${parsed.machine}', not available on this machine (${getMachineId()})` };
+      }
+      try {
+        const { CodexReader } = await import("../integrations/codex/reader.js");
+        const reader = new CodexReader();
+        const win = await reader.getMessageById(parsed.session!, parsed.nativeId, before, after);
+        if (!win) return null;
+        return {
+          type: "opencode_message",
+          message: { id: win.message.id, role: win.message.role, content: win.message.content, timestamp: win.message.timestamp },
+          before: win.before.map(m => ({ id: m.id, role: m.role, content: m.content, timestamp: m.timestamp })),
+          after: win.after.map(m => ({ id: m.id, role: m.role, content: m.content, timestamp: m.timestamp })),
+          session: { id: win.session.id, title: win.session.title, directory: win.session.cwd },
+          source: "codex",
+        };
+      } catch {
+        return null;
+      }
+    }
+
+    case "pi": {
+      if (parsed.machine !== getMachineId()) {
+        return { error: `Message is from machine '${parsed.machine}', not available on this machine (${getMachineId()})` };
+      }
+      try {
+        const { PiReader } = await import("../integrations/pi/reader.js");
+        const reader = new PiReader();
+        const win = await reader.getMessageById(parsed.session!, parsed.nativeId, before, after);
+        if (!win) return null;
+        return {
+          type: "opencode_message",
+          message: { id: win.message.id, role: win.message.role, content: win.message.content, timestamp: win.message.timestamp },
+          before: win.before.map(m => ({ id: m.id, role: m.role, content: m.content, timestamp: m.timestamp })),
+          after: win.after.map(m => ({ id: m.id, role: m.role, content: m.content, timestamp: m.timestamp })),
+          session: { id: win.session.id, title: win.session.title, directory: win.session.cwd },
+          source: "pi",
+        };
+      } catch {
+        return null;
+      }
+    }
+
     case "unknown":
     default: {
       // Backward compat: bare msg_xxx → treat as opencode (no machine qualifier)